FSR sensor compatible with Megadrum?

[jmcdougall]

Yes, I will post! I just looked at an old FSR pad for one of my DrumKats.(scrapped FSR due to 1 bad section) Following the traces I can see a single feed to all 10 areas and individual returns coming back. Looking at a parts DrumKat, I can see that it uses a capacitive termination to ground rather than resistance and the voltage feed looks like it goes to a 4051. My guess is in that vintage (80C31 processor) they fed each of the ten voltages (10 areas on a DrumKat FSR) to a pair of 4051's and then logic switched the 4051's into a single A/D. Over the weekend, I will verify the voltage fed to the pad in the parts machine and then try to test the resistive circuit using the old FSR pad. I am pretty sure i have enough odds and ends of parts to create the inverter circuit. If it looks decent, I might try to dump the output to one of my drums synths.

Dimitri, based on your code, what is the normal voltage range you expect to handle from a piezo -- 0 to ???

[dmitri]

Dimitri, based on your code, what is the normal voltage range you expect to handle from a piezo -- 0 to ???

Vcc for 4051, limits it to +5.

[jmcdougall]

What I meant is what are the typical voltage peaks you see from actual piezo's used in teh more common pads (Roland, Yamaha etc)

[dmitri]

What I meant is what are the typical voltage peaks you see from actual piezo's used in teh more common pads (Roland, Yamaha etc)

Since piezos are high impedance sources, voltage depends on the load. Unloaded they can produce signals in the range of dozens of volts.

[jmcdougall]

OK can I assume that the preferred situation would be a voltage that would scale between 0 and 5V If the 4051 limits at 5V due to saturation then 0 - 5V would provide the best dynamic range as there would be no top end loss of dynamics ?

My next question is how do you determine velocity? Are you watching for the voltage to exceed a threshold and then using fixed timing to go back and re-measure the voltage and convert the relative change in voltage to a velocity? Your methodology will give me a better idea about whether to use a cap to ground or a resistor to ground.

Thanks

Jim

[dmitri]

The velocity is determined by a peak signal.

[jmcdougall]

So you are storing the current and prior values and then doing a compare to see if the voltage drops on the current sample and if it does then use the prior sample as the peak to determine the velocity ?

Back to the earlier question. You remarked that the 4051 is limited to 5V so any peak beyond 5V is treated as a 5v signal ???

I have gotten as far as I can without building an actual circuit. I will be ordering some opamps that I can run on a single ended 5V supply. The same as powers megadrum avoiding a second dual polarity supply. With any luck I will have them by the end of the week so I can breadboard up a prototype over next weekend.

Thanks you you help Dimitri !!

[jmcdougall]

I can say with experience that you can use a FSR with megadrum. I did a little test setup today and was able to trigger a TD-7 without any problems. Based on this, I don't think there would be any issue triggering megadrum. If dimitri or Synthex can look at the LT1006 opamp (Digikey part #LT1006CN8#PBF-ND) and give me some direction as to how to circuit it so that it would generate about 1.5x gain, I will be able to output from 0 - 5V as trigger voltage range. To do this I fed 5V into one side of the FSR (I am using an old FSR from a DumKat that has a couple of bad pads and has a common input). The output is pulled up from ground by a 10K resistor and I pick the input from the junction of the FSR output and the 10K resistor. I am feeding this to the input of an LT1006 opamp (5V singled ended supply and taking the output to the trigger input on the TD-7. The opamp is currently running with unity gain so it is just acting as a buffer and I am getting anout 3V when measuring static pressure. By the end of the end, I should have a storage scope and will be able to calibrate the actual dynamic hits to see what the peaks look like. Although I could have used cheaper opamps, it would have required a dual voltage supply and I wanted to try to keep to something that could be powered by the same 5V supply required by MegaDrum.

[dmitri]

MegaDrum doesn't require sensors to produce full 0-5V range. Thanks to the design and settings MegaDrum can quite happily work with a signal down to 0-1V range.
From a DIYer perspective I'm trying to imagine a pcb with 14 opamps (in case of quad opamps) with all accompanying resistors/diodes/capacitors just to accomodate 56 FSR's. And the proof that FSRs eliminate crosstalk without sacrificing sensitivity yet to be seen.

[jmcdougall]

dimitri
Working with both DrumKats and a TrapKat where all 10(DrumKat) and 21(TrapKAt) FSRs are screened into a single large FSR sheet (TrapKat is physically too big for a single sheet so it uses two), I can attest to the lack of crosstalk yet these systems maintain extremely good sensitivity.

I have two DrumKats that are too old to consider upgrading --- they actually replace the old unit with a new one but its too expensive to consider (about $900) so I only need a board to support the electronics for the 10 pad FSR. These units also have 9 external mono plugs for attaching conventional external pads. It is for this reason that I have been building a MegaDrum. Board is complete, just doing this testing and determining how to mount in a existing DrumKat shell.

However, I could easily see a standard miniboard that would support converting 13 FSRs i.e. 1 octave. Using strip type FSRs, or if I can talk Mario at Alternate Mode into selling the FSR sheets from the MalletKat (should be a standard repair part), you could build the equivalent of a MalletKat. 4 of these boards located under each octave would feed a 56 trigger Megadrum giving you a full 4 octave Marimba/Xylophone style playing surface and still have inputs left over for other external uses.

While i understand that I can adjust the sensitivity in MegaDrum, by not taking advantage of the full 0-5V range, I am exaggerating the dynamic range because smaller voltage changes ie hit pressures are extrapolated i.e your mapping 0-3V into velocity 0-127, so small changes in pressure translate to larger increments of velocity whereas if I use gain to max out to 5V then the mapping is 0-5V into 0-127 so the pad is less sensitive to very small differences in hit pressure. this translates into a feeling of being able to hit from very soft to quite hard and hearing the same thing.

Jim